Annual Report of Benthos, Reef Fish and Invertebrate Surveys for Reef Slope and Reef Flat Areas in Rodrigues 2007

Annual Report of Benthos, Reef Fish and Invertebrate Surveys for Reef Slope and Reef Flat Areas in Rodrigues 2007 E. R. Hardman, F.E. I. Blais, M. S. Desiré, J.S.J Raffin, S. Perrine and Taylor, L. Shoals Rodrigues, Pointe Monier, Rodrigues February 2008

Annual Report of Benthos, Reef Fish and Invertebrate Surveys for Reef Slope and Reef Flat Areas in Rodrigues 2007 E. R. Hardman, F.E.I. Blais, M.S. Desiré, J.S.J Raffin, S. Perrine, and Taylor, L. Shoals Rodrigues, Pointe Monier, Rodrigues Contents 1 Introduction... 1 2 Materials and Methods... 1 3 Results... 3 3.1 Benthos... 3 3.2 Fish... 8 3.3 Invertebrates... 13 3.4 Comparison over 5 years... 18 3.41 Benthos... 18 3.42 Fish... 20 3.43 Invertebrates... 21 3.5 Water Quality... 23 4 Discussion... 25 5 References... 26 6 Appendices... 28 6.1 Transect Locations... 28 6.2 Benthic categories... 32 6.3 Data Tables... 35 List of Figures Figure 1. The location of reef monitoring sites around Rodrigues and the position of the 4 proposed marine reserves..2 Figure 2. The percentage cover of the different benthic habitats at the 6 reef slope sites during (A) Summer (March) and (B) Winter (October)..4 Figure 3. The proportion of coral cover consisting of each of the 9 growth form categories 5 Figure 4. The percentage cover of the different benthic habitats at the 7 reef flat sites during (A) Summer (March) and (B) Winter (October)..6 Figure 5. The proportion of coral cover consisting of each of the 9 growth form categories 7 Figure 6. The distribution of fish families at the 6 reef slope survey sites during the summer survey (March-April).9 Figure 7. The distribution of fish families at the 7 reef flat survey sites during (A) Summer (March) and (B) Winter (October)..12 Figure 8. The distribution of invertebrates at the 6 reef slope survey sites during (A) Summer (March) and (B) Winter (October)..15 Figure 9. The distribution of invertebrates at the 7 reef flat survey sites during (A) Summer (March) and (B) Winter (October)..17 i

Figure 10. Multi-dimensional scaling plot of benthic composition on the reef slope at (A) Rivière Banane, (B) Armand, (C) Grand Bassin and (D) Demi between March 2002 and October 2007. The circles highlight dates with a similar benthic composition.18 Figure 11. Multi-dimensional scaling plot of benthic composition on the reef flat at (A) Rivière Banane, (B) Armand, (C) Grand Bassin (D) Trou Blanc, (E) Cabris and (F) Fous between March 2002 and October 2007. The circles highlight dates with a similar benthic composition..19 Figure 12. Multi-dimensional scaling plot of fish community composition on the reef slope at (A) Rivière Banane, (B) Armand, (C) Grand Bassin and (D) Demi between March 2002 and October 2007. The circles highlight dates with a similar fish community composition 20 Figure 13. Multi-dimensional scaling plot of fish community composition on the reef flat at (A) Rivière Banane, (B) Armand, (C) Grand Bassin (D) Trou Blanc, (E) L Ancre and (F) Cabris between March 2002 and October 2007. The circles highlight dates with a similar fish community composition...21 Figure 14. Multi-dimensional scaling plot of invertebrate community composition on the reef slope at (A) Rivière Banane, (B) Armand, (C) Grand Bassin (D) Demi and (E) North Sables between March 2002 and October 2007. The circles highlight dates with a similar invertebrate community composition..22 Figure 15. Multi-dimensional scaling plot of invertebrate community composition on the reef flat at (A) Rivière Banane, (B) Armand, (C) Grand Bassin (D) Trou Blanc and (E) L Ancre between March 2002 and October 2007. The circles highlight dates with a similar invertebrate community composition..23 Figure 16. The mean nitrate concentration (mg l-1) (±SE) at the 13 survey sites during March/April 2007 24 Figure 17. The mean sea temperature based on continuous temperature loggers at the reef flat sites at Fous, Grand Bassin and Cabris between March 2006 and December 2007 24 List of Tables Table 1. The location of the monitoring sites at which surveys were carried out..1 Table 2. The mean number of coral recruits per settlement tile at each of the 3 surveys sites during summer (October-March) and winter (March-October)..8 Table 3. The mean number of individuals, species, genera and families and species diversity indices for the fish communities at each of the 6 reef slope sites during summer (March) 8 Table 4. The percentage of the 12 key fish species observed, which were below the published length of maturity (FishBase, 2006) at the 6 reef slope sites (RB = Riviere Banane, PA = Armand, GB = Grand Bassin, PD = Demi, IS = North Sables, IF = Ile aux Fous) during Summer (March) and Winter (October) 10 Table 5. The mean number of individuals, species, genera and families and species diversity indices for the fish communities at each of the 7 reef flat sites during summer (March)....11 Table 6. The percentage of the 12 key fish species observed, which were below the published length of maturity (FishBase, 2006) at the 7 reef flat sites (RB = Riviere Banane, PA = Armand, GB = Grand Bassin, TB = Trou Blanc, PlA = L Ancre, IF = Fous, PC = Cabris) during Summer (March) and Winter (October) 13 ii

Table 7. The mean number of individuals and species and species diversity indices for the invertebrate communities at each of the 6 reef slope sites during summer (March) and winter (October).14 Table 8. The mean number of individuals and species and species diversity indices for the invertebrate communities at each of the 7 reef flat sites during summer (March) and winter (October). 16 Acknowledgements This work was funded by the U.K. Department for Environment, Food and Rural Affairs (DEFRA) Darwin Initiative, the North of England Zoological Society (Chester Zoo), WWF and the Fonds Français pour l Environnement Mondial. Thank you to Dr Alasdair Edwards from the University of Newcastle upon Tyne, UK and Dr Charles Anderson from Manta Marine Ltd., Maldives for advice on survey methodology. Shoals Rodrigues would also like to acknowledge the co-operation of the Rodrigues Regional Assembly and the Mauritius Oceanography Institute in the undertaking of this work. iii

Abstract Surveys of reef composition and fish and invertebrate populations were undertaken using the Global Coral Reef Monitoring Network methodology during March/April 2007 and October/November 2007, to continue a monitoring programme that began in 1999. Nine sites around the fringing reef were included: Rivière Banane, Armand, Grand Bassin and Ile aux Fous (reef flat and reef slope stations), Cabris, Trou Blanc and L Ancre (reef flat stations only) and Demi and North Sables (reef slope only). The surveys show that coral cover is high on the reef slopes at Rivière Banane, Grand Bassin and Fous (>45%), but is low at Demi, North Sables and Armand (<30%). In contrast, coral cover was low on all reef flat sites (<30%) indicating that these reefs are being subjected to human and natural impacts. Coral settlement was low, suggesting that these reefs have a limited ability to recover from impacts such as coral bleaching. The fish community at all sites tended to be dominated by Damselfish. Emperors, Snappe, Trevally and Grouper were rare or absent and no Triggerfish, were observed at any site during the surveys. This lack of large piscivorous predators suggests that the fish population may be unbalanced due to overfishing. Fish communities are also dominated by small, under-sized individuals, further indications of overfishing. Invertebrates were low on the reef slope sites and most sites were dominated by the bio-eroding urchin, Echinometra mathaei; this species also dominated the reef flat sites. At all sites, molluscs and crustaceans were either rare or absent; in particular Tridacna clams were in low abundance and large gastropods such as Pleuroploca trapezium were not observed. This may be an indication that local consumption is resulting in overharvesting. It is suggested that the development of marine reserves will protect the healthy reef slopes from future impacts and aim to facilitate recovery of the degraded reef flat areas. iv

1 Introduction Rodrigues is surrounded by a fringing reef, which forms an almost continuous band measuring approximately 90km in length. The reef encloses a shallow lagoon, which, at 240km 2, is twice the area of the island itself. The maximum tidal range is approximately 1.5m, and since the average water depth in the lagoon is less than 2m, many areas are exposed at low spring tides. The water depth immediately beyond the reef slopes is usually within the range of 1 to 3. The island has three major channels, one dredged channel for the main harbour at Port Mathurin in the north, and natural channels in the south near Port Sud Est and in the East at St Francois. Several small passes are also found at intervals around the reef. The reefs of the island consist mainly of scleractinian corals, with Acropora spp. dominant on both the reef flat and reef slope. 140 species of coral were recorded in Rodrigues during the first Marine Biodiversity Workshop held in September 2001, of which 25 were Acropora spp. (Shoals of Capricorn Programme, 2002). The coral cover on the reef slopes around Rodrigues is relatively healthy, while reef flat areas are more heavily impacted, both by fishing (particularly trampling by octopus fishers (Clark, 2001)) and by bleaching events (Hardman et al., 2004). In order to evaluate temporal changes in the health of the coral reef and the populations of fish and invertebrates that it supports, monitoring activities have been carried out since 1999. The data is also made available to the Global Coral Reef Monitoring Network (GCRMN), being submitted via the regional network administered by the Indian Ocean Commission. This report details findings of the GCRMN monitoring activities that took place in 2005, and contains some comparisons with data from 2002-2006 (Hardman et al., 2007; Hardman et al., 2006; Lynch et al., 2004a; Lynch et al., 2004b). 2 Materials and Methods Surveys were carried out at six reef slope and seven reef flat stations around the island by Shoals Rodrigues staff in March/April and October/November 2007. Of these, the stations at Rivière Banane, Grand Bassin, Demi and Cabris were within the 4 proposed marine reserves, whilst the remaining stations were outside of the proposed reserves. The site locations are listed in Table 1 and shown in Figure 1. Table 1. The location of the monitoring sites at which surveys were carried out Reef flat Reef slope Site Name GPS Position GPS Position Rivière Banane 19 40.224 S; 63 28.224 E 19 40.154 S; 63 28.484 E Armand 19 40.272 S; 63 24.982 E 19 40.084 S; 63 24.677 E Grand Bassin 19 39.408 S; 63 21.447 E 19 39.381 S; 63 21.366 E Trou Blanc 19 45.401 S; 63 28.443 E L Ancre 19 º 48.825 S; 63º 25.613 E Demi 19 42.515 S; 63 17.562 E N Sables 19 o 40.770 S; 63 o 17.984 E Cabris 19 o 39.587 S; 63 o 26.406 E Fous 19 o 39.313 S; 63 o 23.593 E 19 o 39.218 E; 63 o 23.977 E 1

N 1km Figure 2. The location of reef monitoring sites around Rodrigues and the position of the 4 proposed marine reserves. The depth on the reef flat was approximately 1m, while surveys on the reef slope took place at between 6m and 12m depth. Reef flat monitoring was carried out by snorkelling, with SCUBA used for reef slope surveys. The location of the site was recorded on a GPS and transects marked by metal bars and buoys. The methodology used was that of the Indian Ocean Commission regional reef monitoring network as described in their manual (Conand et al., 1997), for which three transects were laid at each station. To facilitate replication of the surveys, the transects were permanently marked at, 5m, 1, 15m and 2, using 1m lengths of concrete re-enforcing bar hammered into the reef. Settlement tiles were placed at the reef slope sites at Armand, Rivière Banane and Grand Bassin. Tiles were collected and replaced every 6 months to identify seasonal changes in coral settlement. Counts of all fish encountered were carried out for a belt of 50 x 5m along each transect, by 2 observers while coverage of benthic species and abiotic features was determined by line intercept along the first 2 of each transect. Surveys of invertebrates were carried out by determining abundance of species over a belt 20 x 5m wide for each transect. The lengths of 12 key fish species were also estimated at each site (Anderson, 2006). 2

Permanent temperature loggers were placed at 5 sites to record temperature every hour over a 21-month period (March 2006 December 2007) and nutrient concentration (nitrates, nitrites and phosphates) were determined using the Palintest system. 3 Results 3.1 Benthos During the summer surveys, live hard coral cover on the reef slope was >45% at Grand Bassin, Rivière Banane and Fous, but <40% at Demi, North Sables and Armand (Figure 2). North Sables had a high cover of the soft coral, Sinularia sp. (16%). The site at North of Sables was dominated by coralline algae (54%), whilst Armand and Demi had a high cover of both coralline and turf algae (>20%). All sites had a low percentage of recently dead coral (<2%), rubble (<10%) and macro-algae (<10%). During the winter surveys, live coral cover on the reef slopes was also >45% at Grand Bassin, Rivière Banane and Fous and <30% at Demi, North Sables and Armand. North Sables had a high cover of the soft coral, Sinularia sp. (15%). The site at North of Sables was dominated by coralline algae (54%), whilst Armand and Demi were dominated by turf algae (>35%). Macro-algae was high at Grand Bassin and Fous (>20%), however rubble was low (<5%) and no dead coral was recorded. In both summer and winter surveys, the coral cover on all reef slope sites was dominated by branching Acropora spp., in particular A. austera, A. abrotanoides and A. nobilis (Figure 3). At Grand Bassin, Rivière Banane, Armand and Demi, the corals were dominated by branching Acropora spp. (>50%) with other growth forms forming only a low percentage cover. At Fous and North Sables, branching Acropora spp. constituted over 40% of coral cover and the massive corals Platygyra daedalea and Leptoria phrygia were also common (>25%). 3

(A) Summer (B) Winter Key Living Hard Coral Coralline Algae Zoanthids Soft Coral Macro-Algae Sand Dead Coral Turf Algae Rubble Figure 2. The percentage cover of the different benthic habitats at the 6 reef slope sites during (A) Summer (March) and (B) Winter (October). 4

Key Digitate Acropora Encrusting Acropora Massive Coral Branching Acropora Sub-massive Coral Mushroom Coral Tabular Acropora Encrusting Coral Fire Coral Figure 3. The proportion of coral cover consisting of each of the 9 growth form categories. During the summer surveys, coral cover on the reef flat was 6-14% at Rivière Banane, Armand, Trou Blanc and Fous, 18% at Cabris and 27% at L Ancre (Figure 4). The sites at Rivière Banane, Armand and Fous were dominated by turf algae (>75%), whereas Cabris was dominated by a mixture of turf and coralline algae and the site at Trou Blanc was dominated by the soft corals Xenia sp. and Efflatounaria sp. (54%). Dead coral was only observed at L Ancre (<1%) and rubble was low (<5%) at all sites. No coral bleaching was observed. During the winter surveys, coral cover was 8 14% at Armand and Grand Bassin, 17% at Cabris and 23% at Fous; the remaining 2 sites could not be surveyed due to weather constraints. All 4 sites were dominated by turf algae (>50% cover). Dead coral was not observed during the survey and rubble was low (<1%) at all sites. 5

(A) Summer (B) Winter Key Living Hard Coral Coralline Algae Zoanthids Soft Coral Macro-Algae Sand Dead Coral Turf Algae Rubble Figure 4. The percentage cover of the different benthic habitats at the 7 reef flat sites during (A) Summer (March) and (B) Winter (October). 6

The dominant coral growth forms were more variable on the reef flats than on the reef slope with sub-massive and encrusting growth forms more common (Figure 5). Live hard coral was however dominated by branching Acropora spp. at Armand (A. nobilis and A. abrotanoides), Rivière Banane (A. abrotanoides), Trou Blanc (A. formosa) and L Ancre (A. formosa). Digitate Acropora sp. dominated Grand Bassin (A. digitifera) whereas at Fous and Cabris sub-massive species were dominant (Porites rus and Montipora spumosa). Key Digitate Acropora Foliose Coral Massive Coral Branching Acropora Sub-massive Coral Mushroom Coral Tabular Acropora Encrusting Coral Fire Coral Figure 5. The proportion of coral cover consisting of each of the 9 growth form categories. Coral settlement was low at all sites, and it did not suggest a seasonal recruitment pattern; geographical variations were also difficult to determine. During the period October to March (summer), coral settlement varied from 0 recruits at Rivière Banane to 1.6 recruits per tile at Armand (Table 2). During March to October (winter), coral settlement varied from 0 recruits at Grand Bassin to 1.3 recruits per tile at Rivière Banane. During the winter surveys, 7

recruits were from the family Pocilloporidae and were all approximately 2mm in diameter, suggesting that they had settled approximately 3 months ago. Table 2. The mean number of coral recruits per settlement tile at each of the 3 surveys sites during summer (October-March) and winter (March-October). Rivière Banane Grand Bassin Armand Summer 0.0 0.2 1.6 Winter 1.3 0.0 0.2 3.2 Fish Fish abundance was only recorded during the summer surveys due to logistical reasons. On the reef slopes the highest number of individuals was recorded at Fous (mean of 232 individuals) and Rivière Banane (mean of 229 individuals) and the lowest number at North Ile aux Sables (79 individuals) (Table 3). The highest number of species was observed at Armand (38 species) and the highest number of genera at Rivière Banane (26 genera). The lowest numbers of species, genera and families were recorded at North Sables (23 species, 20 genera and 8 families) and Fous (32 species, 19 genera and 8 families). Species Diversity calculated using the Shannon-Weiner Diversity Index was highest at North Armand (H = 2.473) and lowest at Rivière Banane and Demi (H = 1.878 and H = 1.877 respectively). Similarly, Pielou s Index of Evenness was highest at Armand (J = 0.789) and lowest at Rivière Banane and Demi (J = 0.576 and J = 0.599, respectively). Table 3. The mean number of individuals, species, genera and families and species diversity indices for the fish communities at each of the 6 reef slope sites during summer (March). Rivière Banane Armand Grand Bassin Demi Fous Sables No. Individuals 229.0 130.7 114.3 189.7 232.3 79.0 No. Species 31.0 38.0 32.0 31.0 32.0 23.0 No. Genera 26.0 23.0 21.0 23.0 19.0 20.0 No. Families 11.0 11.0 11.0 9.0 8.0 8.0 H 1.878 2.473 2.173 1.877 2.134 2.153 J 0.576 0.789 0.714 0.599 0.725 0.719 The most commonly observed families were damselfish (Pomacentridae), parrotfish (Scaridae) and surgeonfish (Acanthuridae). Damselfish dominated the reef slopes at all sites except for Grand Bassin and North Sables and were particularly abundant at Rivière Banane (Figure 6). Pomacentridae were dominated by Chromis spp. at Rivière Banane and Demi, Pomcentrus spp. at Armand and Fous and Abudefduf spp. at Grand Bassin and North Sables. Surgeonfish (Acanthuridae) dominated the fish community at Grand Bassin and North Sables and were also common at Armand and Demi. Parrotfish were common at Armand, Grand Bassin, Fous and North Ile aux Sables. The number of Butterflyfish (Chaetodontidae) per site ranged from a mean of 1 to 5 individuals. Emperors (Lethrinidae) were only represented by 1 species (Gnathodentex aurolineatus) and were absent at Demi and North Sables. Lutjanidae were only present at Armand and Grand Bassin and Carangidae at Rivière Banane and Demi, where both families were rare. Groupers (Serranidae) were also rare and no Triggerfish (Balistidae) were recorded at any site. 8

During the summer surveys, the majority of individuals of the 12 key fish species observed were juveniles (Table 4). At Grand Bassin however 100% of Lutjanus fulvus were above the length of maturity; all Mulloidichthys vanicolensis at Fous and all Mulloidichthys flavolineatus at Rivière Banane were adults and all Chaetodon trifasciatus at North Sables were also above the published length of maturity (FishBase, 2006). During the winter surveys, all individuals of Mulloidichthys vanicolensis observed at Rivière Banane were above the published length of maturity, as were the majority of C. trifasciatus, however the majority of Epinephelus spilotoceps individuals were juveniles. Key Acanthuridae Serranidae Pomacentridae Labridae Lethrinidae Scaridae Chaetodontidae Mullidae Other Figure 6. The distribution of fish families at the 6 reef slope survey sites during the summer survey (March-April). 9

Table 4. The percentage of the 12 key fish species observed, which were below the published length of maturity (FishBase, 2006) at the 6 reef slope sites (RB = Riviere Banane, PA = Armand, GB = Grand Bassin, PD = Demi, IS = North Sables, IF = Fous) during Summer (March) and Winter (October). (A) Summer Species Length at maturity (cm) % Juveniles RB PA GB PD IS IF Epinephelus spilotoceps 21.2-0 100 - - 100 Plectropomus laevis 65.2 - - - - - - Lutjanus fulvus 23.8 - - 0 - - - Lethrinus nebulosus 37.4 - - - - - - Mulloidichthys flavolineatus 24.0 0 80 100 - - - Mulloidichthys vanicolensis 24.0 - - - - - 0 Caranx melampygus 51.3 100 - - 100-100 Chaetodon trifasciatus 11.8 33 33 80 33 0 100 Abudefduf sparoides 10.6 - - - - - 100 Scarus ghobban 48.8-100 - - - - Rhinecanthus aculeatus 16.8 - - - - - - Naso unicornis 36.1 100 - - - - - (B) Winter Species Length at maturity (cm) % Juveniles RB PA GB PD IS IF Epinephelus spilotoceps 21.2-100 67 - - 80 Plectropomus laevis 65.2 - - - - - - Lutjanus fulvus 23.8 - - - - - - Lethrinus nebulosus 37.4 - - - - - - Mulloidichthys flavolineatus 24.0-100 - - - - Mulloidichthys vanicolensis 24.0 0 - - - - - Caranx melampygus 51.3 - - - - - 100 Chaetodon trifasciatus 11.8 0 77 0 0 0 33 Abudefduf sparoides 10.6-0 - - - - Scarus ghobban 48.8 - - - - - - Rhinecanthus aculeatus 16.8 - - - 0 - - Naso unicornis 36.1 - - - - 100 - On the reef flat, the highest number of individuals was recorded at L Ancre (mean of 321 individuals) and the lowest number at Riviere Banane (mean of 88 individuals) (Table 5). The highest number of species and families were observed at Fous (26 species, 18 genera and 7 families) and the lowest number of species and families were recorded at Cabris (15 species, 12 genera and 5 families). Species Diversity calculated using the Shannon-Weiner Diversity Index was highest at L Ancre (H = 2.223) and lowest at Trou Blanc (H = 1.288). Pielou s Index of Evenness was highest at Cabris and Pass L Ancre (J = 00.776 and J = 0.769 respectively) and was lowest at Trou Blanc (J = 0.488). 10

Table 5. The mean number of individuals, species, genera and families and species diversity indices for the fish communities at each of the 7 reef flat sites during summer (March) and winter (October). (A) Summer Rivière Banane Armand Trou Blanc L Ancre Cabris Fous No. Individuals 88.0 213.0 296.3 320.7 112.0 254.0 No. Species 20.0 21.0 14.0 26.0 15.0 26.0 No. Genera 15.0 16.0 14.0 18.0 12.0 18.0 No. Families 5.0 6.0 5.0 6.0 5.0 7.0 H 1.731 1.815 1.288 2.223 1.927 1.978 J 0.639 0.655 0.488 0.769 0.776 0.685 Damselfish (Pomacentridae) dominated the reef flat sites at all sites except for Cabris and L Ancre (Figure 7). Pomacentridae were dominated by Stegastes spp. at Rivière Banane, Armand, Rivière Banane and Trou Blanc and by Dascyllus aruanus at L Ancre. The fish community at L Ancre was dominated by parrotfish (small immature parrotfish) whereas surgeonfish (Acanthuridae) and wrasse (Labridae) dominated the fish community at Cabris. A mean of between 0 and 9 butterflyfish were observed at the reef flat sites and no individuals were recorded at Cabris during the summer. Groupers (Serranidae) were only present at Fous and Armand. No Snapper (Lutjanidae), Triggerfish (Balistidae), Trevally (Carangidae) or Fusiliers (Caesionidae) were recorded at any site. During the summer surveys, the majority of individuals of the 12 key fish species observed were juveniles, in particular all Epinephelus spilotoceps individuals observed at all sites were below the published length of maturity (FishBase, 2006) (Table 6). In contrast, Sixty-one percent of Mulloidichthys flavolineatus at Fous and 67% of the same species at Trou Blanc were adults and all Naso unicornis observed at Trou Blanc were adults. Furthermore, over 50% of Chaetodon trifasciatus individuals were adults at Fous, Cabris, L Ancre and Trou Blanc. During winter, all individuals of E. spilotoceps at Fous, Armand and Cabris were juveniles as were all individuals of N. unicornis at Grand Bassin, Fous and Armand. 11

Key Acanthuridae Serranidae Pomacentridae Labridae Lethrinidae Scaridae Chaetodontidae Mullidae Other Figure 7. The distribution of fish families at the 7 reef flat survey sites during (A) Summer (March) and (B) Winter (October). 12

Table 6. The percentage of the 12 key fish species observed, which were below the published length of maturity (FishBase, 2006) at the 7 reef flat sites (RB = Riviere Banane, PA = Armand, GB = Grand Bassin, TB = Trou Blanc, PlA = L Ancre, IF = Fous, PC = Cabris) during Summer (March) and Winter (October). (A) Summer Species Length at maturity (cm) % Juveniles RB PA TB PlA IF PC Epinephelus spilotoceps 21.2-100 100 100 100 100 Plectropomus laevis 65.2 - - - - - - Lutjanus fulvus 23.8 - - - - - - Lethrinus nebulosus 37.4 - - - - - - Mulloidichthys flavolineatus 24.0 - - 33-39 - Mulloidichthys vanicolensis 24.0 - - - - - - Caranx melampygus 51.3 - - - - - - Chaetodon trifasciatus 11.8-100 0 33 47 17 Abudefduf sparoides 10.6 - - - - 0 - Scarus ghobban 48.8-100 - - - - Rhinecanthus aculeatus 16.8 - - - - - - Naso unicornis 36.1-59 0 - - - (B) Winter Species Length at maturity (cm) % Juveniles GB IF PA PC Epinephelus spilotoceps 21.2-100 100 100 Plectropomus laevis 65.2 - - - - Lutjanus fulvus 23.8 - - - - Lethrinus nebulosus 37.4 - - - - Mulloidichthys flavolineatus 24.0-100 - - Mulloidichthys vanicolensis 24.0 - - - - Caranx melampygus 51.3 - - - - Chaetodon trifasciatus 11.8-0 50 0 Abudefduf sparoides 10.6 - - - - Scarus ghobban 48.8 - - - 100 Rhinecanthus aculeatus 16.8 - - - - Naso unicornis 36.1 100 100 100-3.3 Invertebrates The abundance of invertebrates tended to be low on the reef slopes and a total of only 21 species were recorded during all surveys. In both the summer and winter surveys, the highest number of individuals was recorded at North Sables (>600 individuals) and the lowest number of individuals was recorded at Grand Bassin (mean of <10 individuals) (Table 7). The highest number of species was observed at Armand (mean of 10 species in both summer and winter) and the lowest number of species was recorded at North Sables during summer (4 species) and Grand Bassin during winter (3 species). Species Diversity calculated using the Shannon-Weiner Diversity Index was highest at Demi (H = 1.514 in summer and H = 1.615 in winter) and lowest at North Sables (H = 0.033 in summer and H = 0.082 in winter). Pielou s Index of Evenness was highest at Demi in summer (J = 0.728) and at Grand Bassin in summer (J = 0.919) and was lowest at North Ile aux Sables (J = 0.024 in summer and J = 0.046 in winter). 13

Table 7. The mean number of individuals and species and species diversity indices for the invertebrate communities at each of the 6 reef slope sites during summer (March) and winter (October) (A) Summer Rivière Banane Armand Grand Bassin Demi Fous Sables No. Individuals 26.3 137.0 8.6 21.6 183.9 691.1 No. Species 7.0 10.0 7.0 8.0 5.0 4.0 H 1.119 0.452 1.342 1.514 0.147 0.033 J 0.575 0.196 0.690 0.728 0.091 0.024 (B) Winter Rivière Banane Armand Grand Bassin Demi Fous Sables No. Individuals 16.0 252.0 1.3 3.9 134.4 820.5 No. Species 4.0 10.0 3.0 6.0 6.0 6.0 H 1.070 0.408 1.010 1.615 1.572 0.082 J 0.772 0.177 0.919 0.902 0.088 0.046 All sites were dominated by sea urchins (Echinoidea), especially Echinometra mathaei, which was particularly abundant at North Sables, where >650 individuals were recorded (Figure 8). The sea cucumber (Holothuroidea) Stichopus chloronatus was relatively abundant at Armand (mean of 6 individuals), however holothurians were very rare at the other sites. Gastropods were completely absent from Fous in the summer survey and were only found in low numbers at the other sites and the bivalve, Tridacna maxima was only observed at Demi, Armand and Fous (mean of 1 individual at each site). Crustaceans were only represented by hermit crabs (Dardanus sp.) which were observed at Demi, Fous and North Sables. 14

(A) Summer Key Sea Urchins Bivalves Gastropods Sea Cucumbers Octopus Crustaceans Other Echinoderms (B) Winter Figure 8. The distribution of invertebrates at the 6 reef slope survey sites during (A) Summer (March) and (B) Winter (October). 15

Only seventeen invertebrate species were recorded on the 7 reef flat sites during all surveys. The highest number of individuals was recorded at Armand during both the summer and winter surveys (>600 individuals) and the lowest number of individuals was observed at L Ancre in summer (8 individuals) and at Fous during winter (22 individuals) (Table 8). The highest number of species was recorded at Armand and Cabris during summer (10 species) and at Grand Bassin in winter (9 species). The lowest number of species was observed at Fous during both summer and winter (1 species in summer and 3 species in winter). Species Diversity calculated using the Shannon-Weiner Diversity Index was highest at Cabris (H = 0.493 in summer and H = 0.477 in winter) and was lowest at Fous in summer (H = 0.000) and at Armand in winter (H = 0.055). Pielou s Index of Evenness was highest at L Ancre in summer (J = 0.240) and at Fous in winter (J = 0.291) and was lowest at Trou Blanc in summer (J = 0.032) and at Armand in winter (J = 0.031). Table 8. The mean number of individuals and species and species diversity indices for the invertebrate communities at each of the 7 reef flat sites during (A) Summer (March) and (B) Winter (October). (A) Summer Rivière Banane Armand Fous Cabris L Ancre Trou Blanc No. Individuals 533.0 716.7 22.0 169.3 7.6 545.6 No. Species 8.0 10.0 1.0 10.0 2.0 8.0 H 0.330 0.097 0.000 0.493 0.166 0.066 J 0.159 0.042-0.214 0.240 0.032 (B) Winter Grand Bassin Armand Fous Cabris No. Individuals 295.5 608.0 22.4 197.6 No. Species 9.0 6.0 3.0 7.0 H 0.199 0.055 0.320 0.477 J 0.090 0.031 0.291 0.245 All sites except for L Ancre were dominated by Sea Urchins (Echinoidea), especially Echinometra mathaei, which was particularly abundant at Armand, Riviere Banane and Trou Blanc (mean of >500 individuals) (Figure 9). Holothurians were observed at 5 of the sites during the summer surveys (mean of 1-7 individuals per site); however during the winter surveys just 1 individual of Holothuria atra was recorded at Grand Bassin only. The gastropod, Trochus maculatus was relatively abundant at Cabris (mean of 14 individuals in summer and 24 in winter), however molluscs were rare at other sites and completely absent from Fous and Pass L Ancre in the summer surveys. The Bivalve, Tridacna maxima was only observed at Armand, Trou Blanc and Grand Bassin ( 2 individuals). Crustaceans were only represented by hermit crabs (Dardanus sp.) which were present in low numbers (mean of 1 individual per site). 16

(A) Summer Key Sea Urchins Bivalves Gastropods Sea Cucumbers Octopus Crustaceans Other Echinoderms (B) Winter Figure 9. The distribution of invertebrates at the 7 reef flat survey sites during (A) Summer (March) and (B) Winter (October). 17

3.4 Comparison over 5 years 3.41 Benthos Analysis of Similarity (ANOSIM) indicates that on the reef slope, there were significant differences in benthic composition over time at Rivière Banane (R = 0.689, p<0.05), Armand (R = 0.638, p<0.05), Grand Bassin (R = 0.384, p<0.05) and Demi (R = 0.403, p<0.05), however there was no significant difference at Fous or North Sables. At Rivière Banane, SIMPER analysis shows that this was due to very high macro-algae cover in October 2004 (40%) and increased turf algal cover from March 2006 onwards (>20%). These differences are also highlighted by multi-dimensional scaling (MDS) which groups together surveys based on the similarity of their benthic composition (Figure 10). At Armand, March and October 2002 were distinct due to high coral cover (>40%) and there was increased turf algae cover from March 2006 onwards (>15%). Variations in benthic composition at Grand Bassin were less obvious and more governed by variations in coralline algae, with coralline algae low (<20%) in March and October 2002 and high (>25%) in March and October 2003, March 2004 and October 2006; October 2004 was distinct due to high macro-algal cover (36%). There were also no obvious temporal trends at Demi; however October 2004 was distinct due to high macro-algae cover (34%) and there was an increase in turf algae after March 2006 (>25%). Mar-04 Mar-02 Stress: 0.05 High Live Coral Oct-07 Stress: 0.06 Oct-02 Low Turf Algae Oct-02 Mar-03 Oct-03 Mar-02 Oct-06 High Turf Algae Oct-06 Mar-03 Oct-03 Mar-04 High Turf Algae A Oct-04 High Macroalgae Oct-07 B Oct-04 Low Turf Algae Stress: 0.1 Stress: 0 Oct-02 Mar-02 High Macroalgae Low Coralline Algae Oct-04 Mar-04 Mar-03 High Macroalgae Oct-04 Oct-03 Oct-07 Oct-06 C Oct-06 High Coralline Algae Figure 10. Multi-dimensional scaling plot of benthic composition on the reef slope at (A) Rivière Banane, (B) Armand, (C) Grand Bassin and (D) Demi between March 2002 and October 2007. The circles highlight dates with a similar benthic composition. D Oct-07 High Turf Algae On the reef flat, ANOSIM indicates that there were significant difference in benthic composition over time at all 7 sites (p<0.05). At Rivière Banane, SIMPER analysis shows that this was due to high macro-algae (17%) in March 2003 and high turf algae (>70%) in October 2003, March 2006 and March 2007. These differences are also highlighted by multi- 18

dimensional scaling (MDS) which groups together surveys based on the similarity of their benthic composition (Figure 11). At Armand, March 2002 and October 2005 were distinct due to high dead coral (>30%) and turf algae was high (>70%) from March 2006 onwards. Similarly, at Grand Bassin, March 2002 was distinct due to high dead coral (26%) and turf algae was high (>25%) from March 2006 onwards. At Trou Blanc, October 2002 was distinct due to high macro-algae (37%) and October 2005 was distinct due to high dead coral (32%) and turf algal cover was high from March 2006 onwards. At the remaining sites, there were no obvious temporal trends and variations tended to be due to changes in turf and coralline algal cover, however October 2005 was distinct at Fous due to high dead coral (38%). Mar-02 Oct-04 Mar-04 Oct-03 Stress: 0.01 Low Turf Algae Oct-03 Mar-03 Stress: 0.08 Oct-02 Low Turf Algae High Turf Algae High Turf Algae Oct-06 Mar-04 Oct-04 Oct-07 Oct-02 A High Macroalgae Mar-03 B High Dead Coral Mar-02 Stress: 0.05 Stress: 0.07 Oct-07 Oct-04 Low Turf Algae High Turf Algae Oct-04 Oct-06 High Turf Algae Mar-04 Oct-06 Mar-02 Oct-02 High Macroalgae Mar-03 Oct-03 Mar-03 Mar-04 Oct-03 C Oct-02 High Dead Coral Mar-02 D High Dead Coral Stress: 0 Stress: 0 High Turf Algae Oct-07 High Turf Algae Oct-07 Low Turf Algae Oct-06 High Dead Coral E Figure 11. Multi-dimensional scaling plot of benthic composition on the reef flat at (A) Rivière Banane, (B) Armand, (C) Grand Bassin (D) Trou Blanc, (E) Cabris and (F) Fous between March 2002 and October 2007. The circles highlight dates with a similar benthic composition. F 19

3.42 Fish On the reef slope, ANOSIM indicates that there were significant differences in fish community composition over time at all sites (p<0.05). SIMPER analysis however does not highlight any temporal trends, but indicates that these differences are due to the occasional occurrence of large shoals of fish such as Pterocaesio tile (Caesionidae) in March 2003 and 2006 (Rivière Banane, Grand Bassin and Demi), Gnathodentex aureolineatus (Lethrinidae) in March and October 2002 and Acanthurus sp. in October 2002 ( Armand), October 2003 (Grand Bassin) and March 2007 ( Demi). These variations are also highlighted by multi-dimensional scaling (MDS) which groups together surveys based on the similarity of their fish community composition (Figure 12). Stress: 0.05 High Acanthuridae Oct-02 Stress: 0.05 Mar-04 Mar-02 Mar-02 Oct-02 Oct-03 Oct-06 Mar-04 Oct-04 Mar-03 High Caesionidae Oct-03 Oct-04 A High Scaridae B Mar-03 High Caesionidae Oct-06 High Lethrinidae Oct-02 Stress: 0.04 Stress: 0 Mar-02 High Caesionidae Mar-03 High Acanthuridae Oct-04 Oct-06 Oct-03 Mar-04 Oct-04 High Acanthuridae C Oct-06 Figure 12. Multi-dimensional scaling plot of fish community composition on the reef slope at (A) Rivière Banane, (B) Armand, (C) Grand Bassin and (D) Demi between March 2002 and October 2007. The circles highlight dates with a similar fish community composition. D On the reef flat, ANOSIM indicates that there were significant differences in fish community composition over time at all 7 sites (p<0.05). SIMPER analysis however does not highlight any temporal trends, but indicates that these differences are due to variations in Acanthuridae (Riviere Banane, Trou Blanc and L Ancre) and shoals of Mullidae during March 2002 ( Armand) and Lethrinidae (Grand Bassin). These variations are also highlighted by multi-dimensional scaling (MDS) which groups together surveys based on the similarity of their fish community composition (Figure 13). 20

Stress: 0.02 Stress: 0.08 High Acanthuridae Oct-03 Mar-02 No Acanthuridae High Mullidae Mar-03 Mar-02 Mar-04 Oct-04 Oct-02 Oct-04 Mar-04 Oct-03 Mar-03 Oct-02 High Pomacentridae Oct-06 A B Oct-03 Oct-02 Mar-03 Stress: 0.05 Mar-02 High Lethrinidae High Scaridae Mar-02 Mar-04 No Acanthuridae Stress: 0.07 Oct-06 Oct-06 Mar-04 Oct-04 Oct-04 C Low Pomacentridae D Mar-03 Oct-02 No Chaetodontidae Oct-03 Stress: 0.09 Stress: 0 Oct-04 Oct-06 Oct-06 Oct-03 Mar-03 E Low Acanthuridae Mar-04 Figure 13. Multi-dimensional scaling plot of fish community composition on the reef flat at (A) Rivière Banane, (B) Armand, (C) Grand Bassin (D) Trou Blanc, (E) L Ancre and (F) Cabris between March 2002 and October 2007. The circles highlight dates with a similar fish community composition. F 3.43 Invertebrates On the reef slope, ANOSIM indicates that there were significant differences in invertebrate community composition over time at all sites except for Fous (p<0.05). SIMPER analysis indicates that at Rivière Banane and Grand Bassin this is due to the absence of Diadema sp. from March 2005 onwards. At Armand the difference is due to an increase in the urchin, Echinometra mathaei from October 2005 onwards, whereas at Demi it is due to the absence of Featherstars in October 2004, 2005 and 2007 and at Nprth Sables the difference is due to lower numbers of E. mathaei in March and October 2006. These variations are also highlighted by multi-dimensional scaling (MDS) which groups together surveys based on the similarity of their invertebrate community composition (Figure 14). No Chaetodontidae 21

No Diadema sp. Stress: 0.09 Mar-04 Stress: 0.08 Mar-04 Oct-02 Mar-02 Low E. mathaei. High E. mathaei. Oct-02 Oct-06 Oct-07 Oct-03 Oct-06 Mar-02 Oct-03 Mar-03 Mar-03 Oct-04 Oct-04 A B Oct-07 No Diadema sp. Oct-03 Stress: 0.11 Presence of Featherstars Stress: 0.01 Oct-06 Diadema sp. Oct-06 Oct-04 Mar-04 Oct-02 Mar-02 Oct-07 Oct-04 Mar-03 Oct-07 C D Stress: 0 Oct-07 Oct-06 Lower E. mathaei. E High E. mathaei. Figure 14. Multi-dimensional scaling plot of invertebrate community composition on the reef slope at (A) Rivière Banane, (B) Armand, (C) Grand Bassin (D) Demi and (E) North Sables between March 2002 and October 2007. The circles highlight dates with a similar invertebrate community composition. On the reef flat, ANOSIM indicates that there were significant differences in invertebrate community composition over time at 5 of the sites (p<0.05). SIMPER analysis indicates that at Rivière Banane, Armand, Grand Bassin and Trou Blanc this is due to an increase in the abundance of Echinometra mathaei from October 2005 onwards. At L Ancre, the difference is due to a seasonal variation in invertebrate abundance, with a low abundance of all invertebrates observed during the summer surveys (March). These variations are also highlighted by multi-dimensional scaling (MDS) which groups together surveys based on the similarity of their invertebrate community composition (Figure 15). 22

Oct-04 Lower E. mathaei. Stress: 0.03 High E. mathaei. Stress: 0.04 Mar-02 Oct-02 Oct-07 Oct-06 Oct-03 Mar-04 Mar-02 Oct-04 Mar-04 Oct-03 Mar-03 A Oct-02 High E. mathaei. B Lower E. mathaei. Mar-03 Stress: 0.02 Stress: 0.02 Diadema sp. Lower E. mathaei. High E. mathaei. Oct-04 Oct-02 Oct-06 Oct-02 Oct-04 Mar-02 Mar-04 High E. mathaei. Mar-04 Oct-06 Mar-03 Oct-03 Mar-02 Mar-03 Oct-03 C Lower E. mathaei. D Stress: 0 Oct-04 Low invertebrate abundance. Mar-04 Oct-06 Oct-03 E Figure 15. Multi-dimensional scaling plot of invertebrate community composition on the reef flat at (A) Rivière Banane, (B) Armand, (C) Grand Bassin (D) Trou Blanc and (E) L Ancre between March 2002 and October 2007. The circles highlight dates with a similar invertebrate community composition. 3.5 Water Quality Nitrite and phosphate levels at all 13 sites were negligible. Nitrate concentration was however significantly higher at the reef flat sites at Armand and Cabris than at the remaining sites (1-way ANOVA, p<0.001) (Figure 16). Sea surface temperature measured at the 3 reef flat sites, varied between 19.82 o C in August 2007 and 31.59 o C at Grand Bassin. Mean temperature was significantly higher at Cabris that at Grand Bassin and Fous (1-way ANOVA, p<0.001) (Figure 17). 23

Figure 16. The mean nitrate concentration (mg l-1) (±SE) at the 13 survey sites during March/April 2007. Figure 17. The mean sea temperature based on continuous temperature loggers at the reef flat sites at Fous, Grand Bassin and Cabris between March 2006 and December 2007. 24

4 Discussion Hard coral cover was high on the reef slope at Rivière Banane, Grand Bassin and Fous (>45%), but was low at Demi, North Sables and Armand, which were dominated by coralline and turf algae. There was however low dead coral cover at all sites, suggesting that the sites are generally healthy. Faure (1982) also notes that coral cover on the reef slopes was 50-60%, suggesting that there has been little change over time. There were increases in the red macro-algae, Asparagopsis taxiformis, at some sites during October 2004, however macro-algae returned to low levels in the following summer surveys, suggesting this may be natural. Indeed, Faure (1982) also comments that this species was common on sheltered reefs in the north of the island. In contrast, coral cover was low on the reef flat sites (<30%). Faure (1982) and Faure and Montaggioni (1974) observed that live coral cover was 2-30% on the reef flats, suggesting that there has been some decline in coral cover over time, particular at the northern sites. There were also temporal increases in dead coral cover, with high values recorded during March 2002 at Armand and Grand Bassin and during October 2005 at Armand, Trou Blanc and Fous. Coral bleaching affected corals on the reef flats during 2002 (Hardman et al., 2004) and 2005 (Hardman et al., in press), causing 90% bleaching of the dominant branching Acropora species and this is likely to have resulted in the increase in dead coral observed at these reef flat sites. Sea temperature was highest at the sheltered inshore site of Cabris, suggesting that corals at this site are very vulnerable to future bleaching events. Coral settlement was also low, with <2 recruits observed per tile and no recruits observed at Rivière Banane during summer or Grand Bassin during winter, suggesting that these reefs have a limited ability to recover from impacts such as coral bleaching. On the reef slope all sites except Grand Basin and North Sables were dominated by Damselfish, with very high numbers being recorded at Rivière Banane. Damselfish were also dominant at 5 of the reef flat sites. At all sites, Emperors, Snapper, Trevally and Grouper were rare or absent. This lack of large piscivorous predators suggests that the fish population may be unbalanced due to overfishing. Sites showed no obvious temporal variations in community composition and differences tended to be due to the presence of occasional large shoals of surgeonfish and fusiliers. At both the reef slope and reef flat sites, the majority of key fish species observed were below the published length of maturity. This suggests that either these areas are important grounds for juveniles or that the population is dominated by small undersized individuals due to intense fishing of the adult individuals. Studies of the lagoon fishery in Rodrigues also indicate overfishing due to a decline in fish catches over time (Hardman et al., 2007) and a decline in the catch of predatory species such as Lethrinus harak (Pearson, 1988). Invertebrates were low on the reef slope sites and all sites, were dominated by the urchin, Echinometra mathaei. This species also dominated all of the reef flat sites, except for L Ancre. There were temporal variations in the abundance of E. mathaei on the reef slope at Armand and North sables and there has been a general increase in numbers from October 2005 onwards on the reef flat at Rivière Banane, Armand, Grand Bassin and Trou Blanc. The high number of this species may be as a result of intense fishing pressure and the removal of predator fish species. E. mathaei is a bio-eroder and so their high density is cause for concern, especially with possible increase in the incidences of coral bleaching resulting in higher coral mortality. At all sites molluscs and crustaceans were either rare or absent; in particular Tridacna clams were in low abundance and large gastropods such as 25

Pleuroploca trapezium were not observed. This may be an indication that local consumption is resulting in over-harvesting. Nutrient values were low, however nitrate values were significantly higher at Armand and Cabris. These sites are the two closest sites to Port Mathurin and this suggests that the lagoon is starting to become polluted through domestic and industrial run-off from the main town. Further surveys carried out at different times of the year are however needed to confirm whether this is indeed the case. Considering the sites within the proposed marine reserves, it can be seen that the reef slopes are generally healthy with high coral cover (>45%) at Grand Bassin and Rivière Banane and a low percentage cover of dead coral and rubble. Fish communities are dominated by Damselfish, with Surgeonfishfish dominant at Grand Bassin and the endemic Damselfish, Pomacentrus rodriguesensis, was observed at Rivière Banane. At present, the majority of fish species are undersized, however adult individuals of the commercially important species, Caranx melampygus, Naso unicornis and Mulloidichthys flavolineatus were observed. Invertebrates tend to be low at all sites. The reef flat sites within the proposed marine reserves are more degraded with low coral cover (<20%). The abundance of fish was low at Rivière Banane and at Cabris and the majority of individuals were juveniles. The invertebrate community at these sites was dominated by very high numbers of the urchin Echinometra mathaei, suggesting that erosion of the reef structure may be occurring. The presence of marine reserves in these areas will protect the healthy reef slopes from future impacts and aim to facilitate recovery of the degraded reef flat areas. 5 References Anderson, R. C. (2006). Impacts of marine reserves in Rodrigues; report of a training visit to Shoals Rodrigues, March 2006. Unpublished report 21pp. Clark T.H. (2001). The Status of the Coral Reefs in Rodrigues. Shoals of Capricorn Programme unpublished report. Conand C., Chabanet P., Quod J.P and Bigot L. (1997). Guidelines for coral reef monitoring in the south-west region of the Indian Ocean. Indian Ocean Commission. Faure, G. (1982). Recherche sur les peuplements de Scleratiniaires des récifs coralliens de l Archipel des Mascareignes. PhD Thesis, Université d Aix-Marseille, France. Faure, G. and Montaggioni, L. (1974). Le récif corallien de l Ile Rodrigue (Archipel des Mascareigens, Océan Indien). Géomorphologie et répartition des peuplements. Journal of the Marine Biological Association of India 16: 1-30. Hardman, E. R., Meunier, M. S., Turner, J. R., Lynch, T. L., Taylor, M and Klaus, R. (2004). The extent of coral bleaching in Rodrigues, 2002. Journal of Natural History 38: 3077-3089. Hardman, E. R., Blais, F. E. I., Desiré, S. M., Raffin, J. S. J., Perrine, S., Raffaut, R. and Chinien-Chetty, M. (2006). Annual report of benthos, reef fish and invertebrate surveys for reef slope and reef flat areas in Rodrigues 2005. Shoals Rodrigues report, 43pp. Hardman, E. R., Blais, F. E. I., Desiré, S. M., Raffin, J. S. J., Perrine, S., Chinien-Chetty, M. and Towill, J. (2007). Annual report of benthos, reef fish and invertebrate surveys for reef slope and reef flat areas in Rodrigues 2006. Shoals Rodrigues report, 51pp. Hardman, E. R., Stampfli, N. S., Hunt, L., Perrine, S., Perry, A. and Raffin, J. S. J. (in press). The impacts of coral bleaching in Rodrigues, western Indian Ocean. Atoll Research Bulletin. 26

Lynch T.L., Meunier M.S., Hooper T.E.J., Blais F.E.I., Raffin J.S.J, Perrine S., Félicité N., Lisette J., Grandcourt J.W. (2002). Annual report of benthos, reef fish and invertebrate surveys for Rodrigues 2002. Shoals Rodrigues report, 30pp. Lynch, T. L., Meunier, M. S., Hooper, T. E. J., Blais, F. E. I., Raffin, J. S. J., Perrine, S., and Raffaut, R. (2004a). Annual report of benthos, reef fish and invertebrate surveys for reef slope and reef flat areas in Rodrigues 2003. Shoals Rodrigues report, 52pp. Lynch, T. L., Meunier, M. S., Blais, F. E. I., Raffin, J. S. J., Perrine, S., Raffaut, R. and Perry, A. C. (2004b). Annual report of benthos, reef fish and invertebrate surveys for reef slope and reef flat areas in Rodrigues 2004. Shoals Rodrigues report, 48pp. Pearson, M. P. (1988). Rodrigues. Rapid survey of the status of the exploitation and environmental damage of the lagoon and coral reefs of Rodrigues. Report prepared for the project Assistance to artisanal fishermen and development of outer reef fishery. FAO, Rome. Shoals of Capricorn Programme (2002). Rodrigues Marine Biodiversity Workshop 10 September to 05 October 2001 Final Report. 27

6 Appendices 6.1 Transect Locations Reef Edge Sand T3 T2 T1 6 7m depth 5 5 Sand 5 Rivière Banane Reef Slope Reef Edge T3 T2 T1 6 7m depth 5 5 5 Armand Reef Slope Reef Edge 5 T3 5 T2 Gully T1 5 11 12m depth Demi Reef Flat 28

Reef Edge 5 5 5 T3 T2 T1 7 9m depth Sand Grand Bassin Reef Slope Reef Edge T1 T2 T3 8 9m depth 5 5 5 Fous Reef Slope Reef Edge 5 5 T1 T2 North Sables Reef Slope T3 5 9-11m depth 29

5 Reef Edge 5 T2 5 T3 T1 Rivière Banane Reef Flat 5 Reef Edge 5 T1 5 T3 T2 Armand Reef Flat Reef Edge 5 T1 T2 5 T3 5 Grand Bassin Reef Flat 5 Reef Edge 5 5 T3 T2 T1 Trou Blanc Reef Flat 30

5 5 5 Reef Edge T3 T2 T1 L Ancre Reef Flat Reef Edge T1 5 T2 5 T3 5 Cabris Reef Flat T1 5 T2 5 T3 Reef Edge 5 Fous Reef Flat 31

6.2 Benthic categories Acropora branching (ACB) Acropora digitate (ACD) Acropora encrusting (ACE) Acropora tabular (ACT) Coral branching (CB) Coral encrusting (CE) Coral foliose (CF) Coral mushroom (CL) 32

Coral massive (CS) Coral sub-massive (CS) Millepora (MI) Soft coral (SC) Zoanthids (ZO) Dead coral (DC) Rubble (RB) Macroalgae (MA) 33

Turf algae (TA) Coralline algae (CA) 34